Related Experiment Video
Updated: Apr 16, 2026
13:45
Hybrid µCT-FMT imaging and image analysis
Published on: June 4, 2015
13.8K
Adaptive Mesh Expansion Model (AMEM) for liver segmentation from CT image
Xuehu Wang1, Jian Yang1, Danni Ai1
1Beijing Engineering Research Center of Mixed Reality and Advanced Display, School of Optics and Electronics, Beijing Institute of Technology, Beijing 100081, China.
Plos One
|March 14, 2015
Summary
A new adaptive mesh expansion model (AMEM) accurately segments livers in CT scans using a deformable simplex model. This method precisely controls internal and external forces, improving accuracy, especially at sharp corners, outperforming existing algorithms.
Area of Science:
- Medical Imaging
- Computer-Aided Diagnosis
- Computational Anatomy
Background:
- Accurate liver segmentation in computed tomography (CT) images is crucial for diagnosis and treatment planning.
- Existing segmentation methods often struggle with irregular shapes and sharp boundaries of the liver.
Purpose of the Study:
- To introduce a novel adaptive mesh expansion model (AMEM) for robust and accurate liver segmentation from CT images.
- To enhance segmentation precision by adaptively refining mesh components for complex anatomical structures.
Main Methods:
- Utilized a virtual deformable simplex model (DSM) to represent the mesh, allowing for precise vertex manipulation.
- Integrated balloon, edge, and gradient forces for external driving forces and tangential/normal forces for internal smoothness control.
- Implemented adaptive decomposition of triangular facets to improve accuracy at irregularly sharp liver corners.
Main Results:
- The AMEM algorithm demonstrated superior performance compared to six other state-of-the-art methods on 10 clinical datasets.
- Achieved a mean overlap error of 6.8% and a mean volume difference of 2.7%.
- Attained an average symmetric surface distance of 1.3 mm and a root mean square symmetric surface distance of 2.7 mm.
Conclusions:
- The proposed AMEM is an effective and robust method for liver segmentation in CT images.
- The adaptive mesh refinement and controlled deformable forces contribute to high segmentation accuracy.
- AMEM shows significant potential for clinical applications requiring precise liver volume assessment.
